DouglasBraga's picture
End of training
18b86e4 verified
raw
history blame
56 kB
{
"best_metric": 0.8892018779342723,
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-leukemia-08-2024.v1.1\\checkpoint-2187",
"epoch": 9.984,
"eval_steps": 500,
"global_step": 3120,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.032,
"grad_norm": 2.98797607421875,
"learning_rate": 1.6025641025641025e-06,
"loss": 0.712,
"step": 10
},
{
"epoch": 0.064,
"grad_norm": 6.277791976928711,
"learning_rate": 3.205128205128205e-06,
"loss": 0.6981,
"step": 20
},
{
"epoch": 0.096,
"grad_norm": 5.27135705947876,
"learning_rate": 4.807692307692308e-06,
"loss": 0.6783,
"step": 30
},
{
"epoch": 0.128,
"grad_norm": 6.831840991973877,
"learning_rate": 6.41025641025641e-06,
"loss": 0.6498,
"step": 40
},
{
"epoch": 0.16,
"grad_norm": 8.679888725280762,
"learning_rate": 8.012820512820515e-06,
"loss": 0.6079,
"step": 50
},
{
"epoch": 0.192,
"grad_norm": 11.260509490966797,
"learning_rate": 9.615384615384616e-06,
"loss": 0.579,
"step": 60
},
{
"epoch": 0.224,
"grad_norm": 17.14796257019043,
"learning_rate": 1.1217948717948719e-05,
"loss": 0.5673,
"step": 70
},
{
"epoch": 0.256,
"grad_norm": 10.239635467529297,
"learning_rate": 1.282051282051282e-05,
"loss": 0.5263,
"step": 80
},
{
"epoch": 0.288,
"grad_norm": 7.6730055809021,
"learning_rate": 1.4423076923076923e-05,
"loss": 0.5371,
"step": 90
},
{
"epoch": 0.32,
"grad_norm": 7.861423492431641,
"learning_rate": 1.602564102564103e-05,
"loss": 0.5144,
"step": 100
},
{
"epoch": 0.352,
"grad_norm": 9.051122665405273,
"learning_rate": 1.762820512820513e-05,
"loss": 0.5046,
"step": 110
},
{
"epoch": 0.384,
"grad_norm": 14.384810447692871,
"learning_rate": 1.923076923076923e-05,
"loss": 0.4784,
"step": 120
},
{
"epoch": 0.416,
"grad_norm": 12.508864402770996,
"learning_rate": 2.0833333333333336e-05,
"loss": 0.4626,
"step": 130
},
{
"epoch": 0.448,
"grad_norm": 14.830451011657715,
"learning_rate": 2.2435897435897437e-05,
"loss": 0.4459,
"step": 140
},
{
"epoch": 0.48,
"grad_norm": 52.42909622192383,
"learning_rate": 2.4038461538461542e-05,
"loss": 0.4163,
"step": 150
},
{
"epoch": 0.512,
"grad_norm": 32.466487884521484,
"learning_rate": 2.564102564102564e-05,
"loss": 0.4801,
"step": 160
},
{
"epoch": 0.544,
"grad_norm": 30.3638973236084,
"learning_rate": 2.724358974358974e-05,
"loss": 0.4896,
"step": 170
},
{
"epoch": 0.576,
"grad_norm": 27.437076568603516,
"learning_rate": 2.8846153846153845e-05,
"loss": 0.4598,
"step": 180
},
{
"epoch": 0.608,
"grad_norm": 11.979604721069336,
"learning_rate": 3.0448717948717947e-05,
"loss": 0.421,
"step": 190
},
{
"epoch": 0.64,
"grad_norm": 68.23617553710938,
"learning_rate": 3.205128205128206e-05,
"loss": 0.4518,
"step": 200
},
{
"epoch": 0.672,
"grad_norm": 41.465309143066406,
"learning_rate": 3.365384615384616e-05,
"loss": 0.4403,
"step": 210
},
{
"epoch": 0.704,
"grad_norm": 14.963610649108887,
"learning_rate": 3.525641025641026e-05,
"loss": 0.4282,
"step": 220
},
{
"epoch": 0.736,
"grad_norm": 16.748451232910156,
"learning_rate": 3.685897435897436e-05,
"loss": 0.3553,
"step": 230
},
{
"epoch": 0.768,
"grad_norm": 47.92767333984375,
"learning_rate": 3.846153846153846e-05,
"loss": 0.4068,
"step": 240
},
{
"epoch": 0.8,
"grad_norm": 32.47563552856445,
"learning_rate": 4.006410256410257e-05,
"loss": 0.4356,
"step": 250
},
{
"epoch": 0.832,
"grad_norm": 32.39358901977539,
"learning_rate": 4.166666666666667e-05,
"loss": 0.3873,
"step": 260
},
{
"epoch": 0.864,
"grad_norm": 17.5948429107666,
"learning_rate": 4.326923076923077e-05,
"loss": 0.3428,
"step": 270
},
{
"epoch": 0.896,
"grad_norm": 34.92612075805664,
"learning_rate": 4.4871794871794874e-05,
"loss": 0.4007,
"step": 280
},
{
"epoch": 0.928,
"grad_norm": 12.984992980957031,
"learning_rate": 4.6474358974358976e-05,
"loss": 0.4148,
"step": 290
},
{
"epoch": 0.96,
"grad_norm": 62.389808654785156,
"learning_rate": 4.8076923076923084e-05,
"loss": 0.3258,
"step": 300
},
{
"epoch": 0.992,
"grad_norm": 20.061573028564453,
"learning_rate": 4.9679487179487185e-05,
"loss": 0.32,
"step": 310
},
{
"epoch": 0.9984,
"eval_accuracy": 0.6910798122065728,
"eval_loss": 0.9902069568634033,
"eval_runtime": 15.0205,
"eval_samples_per_second": 70.903,
"eval_steps_per_second": 2.264,
"step": 312
},
{
"epoch": 1.024,
"grad_norm": 95.87650299072266,
"learning_rate": 4.985754985754986e-05,
"loss": 0.4095,
"step": 320
},
{
"epoch": 1.056,
"grad_norm": 15.716072082519531,
"learning_rate": 4.9679487179487185e-05,
"loss": 0.3434,
"step": 330
},
{
"epoch": 1.088,
"grad_norm": 24.56471061706543,
"learning_rate": 4.95014245014245e-05,
"loss": 0.3115,
"step": 340
},
{
"epoch": 1.12,
"grad_norm": 27.441110610961914,
"learning_rate": 4.932336182336182e-05,
"loss": 0.3818,
"step": 350
},
{
"epoch": 1.152,
"grad_norm": 19.63530921936035,
"learning_rate": 4.9145299145299147e-05,
"loss": 0.4001,
"step": 360
},
{
"epoch": 1.184,
"grad_norm": 59.710182189941406,
"learning_rate": 4.896723646723647e-05,
"loss": 0.3836,
"step": 370
},
{
"epoch": 1.216,
"grad_norm": 11.739165306091309,
"learning_rate": 4.878917378917379e-05,
"loss": 0.3505,
"step": 380
},
{
"epoch": 1.248,
"grad_norm": 9.130122184753418,
"learning_rate": 4.8611111111111115e-05,
"loss": 0.2724,
"step": 390
},
{
"epoch": 1.28,
"grad_norm": 35.71299362182617,
"learning_rate": 4.8433048433048433e-05,
"loss": 0.2992,
"step": 400
},
{
"epoch": 1.312,
"grad_norm": 23.9375,
"learning_rate": 4.825498575498576e-05,
"loss": 0.3046,
"step": 410
},
{
"epoch": 1.3439999999999999,
"grad_norm": 25.04848289489746,
"learning_rate": 4.8076923076923084e-05,
"loss": 0.315,
"step": 420
},
{
"epoch": 1.376,
"grad_norm": 18.383527755737305,
"learning_rate": 4.78988603988604e-05,
"loss": 0.3109,
"step": 430
},
{
"epoch": 1.408,
"grad_norm": 13.666149139404297,
"learning_rate": 4.772079772079772e-05,
"loss": 0.2999,
"step": 440
},
{
"epoch": 1.44,
"grad_norm": 9.764185905456543,
"learning_rate": 4.7542735042735045e-05,
"loss": 0.2615,
"step": 450
},
{
"epoch": 1.472,
"grad_norm": 15.155381202697754,
"learning_rate": 4.736467236467237e-05,
"loss": 0.261,
"step": 460
},
{
"epoch": 1.504,
"grad_norm": 14.417835235595703,
"learning_rate": 4.718660968660969e-05,
"loss": 0.2596,
"step": 470
},
{
"epoch": 1.536,
"grad_norm": 26.80813217163086,
"learning_rate": 4.700854700854701e-05,
"loss": 0.3011,
"step": 480
},
{
"epoch": 1.568,
"grad_norm": 29.056140899658203,
"learning_rate": 4.683048433048433e-05,
"loss": 0.2741,
"step": 490
},
{
"epoch": 1.6,
"grad_norm": 8.897234916687012,
"learning_rate": 4.665242165242166e-05,
"loss": 0.2906,
"step": 500
},
{
"epoch": 1.6320000000000001,
"grad_norm": 15.488897323608398,
"learning_rate": 4.6474358974358976e-05,
"loss": 0.3148,
"step": 510
},
{
"epoch": 1.6640000000000001,
"grad_norm": 45.335018157958984,
"learning_rate": 4.62962962962963e-05,
"loss": 0.3026,
"step": 520
},
{
"epoch": 1.696,
"grad_norm": 8.628056526184082,
"learning_rate": 4.611823361823362e-05,
"loss": 0.2719,
"step": 530
},
{
"epoch": 1.728,
"grad_norm": 11.930350303649902,
"learning_rate": 4.594017094017094e-05,
"loss": 0.2831,
"step": 540
},
{
"epoch": 1.76,
"grad_norm": 7.246833324432373,
"learning_rate": 4.576210826210827e-05,
"loss": 0.2292,
"step": 550
},
{
"epoch": 1.792,
"grad_norm": 31.572784423828125,
"learning_rate": 4.558404558404559e-05,
"loss": 0.2698,
"step": 560
},
{
"epoch": 1.8239999999999998,
"grad_norm": 9.255568504333496,
"learning_rate": 4.5405982905982906e-05,
"loss": 0.2469,
"step": 570
},
{
"epoch": 1.8559999999999999,
"grad_norm": 13.890963554382324,
"learning_rate": 4.522792022792023e-05,
"loss": 0.2614,
"step": 580
},
{
"epoch": 1.888,
"grad_norm": 13.776374816894531,
"learning_rate": 4.504985754985755e-05,
"loss": 0.264,
"step": 590
},
{
"epoch": 1.92,
"grad_norm": 7.997173309326172,
"learning_rate": 4.4871794871794874e-05,
"loss": 0.2548,
"step": 600
},
{
"epoch": 1.952,
"grad_norm": 44.6745719909668,
"learning_rate": 4.46937321937322e-05,
"loss": 0.2476,
"step": 610
},
{
"epoch": 1.984,
"grad_norm": 13.378036499023438,
"learning_rate": 4.451566951566952e-05,
"loss": 0.2625,
"step": 620
},
{
"epoch": 2.0,
"eval_accuracy": 0.7690140845070422,
"eval_loss": 0.5525702238082886,
"eval_runtime": 14.9955,
"eval_samples_per_second": 71.021,
"eval_steps_per_second": 2.267,
"step": 625
},
{
"epoch": 2.016,
"grad_norm": 7.805148601531982,
"learning_rate": 4.4337606837606836e-05,
"loss": 0.2232,
"step": 630
},
{
"epoch": 2.048,
"grad_norm": 10.30744457244873,
"learning_rate": 4.415954415954416e-05,
"loss": 0.2577,
"step": 640
},
{
"epoch": 2.08,
"grad_norm": 15.62030029296875,
"learning_rate": 4.3981481481481486e-05,
"loss": 0.2273,
"step": 650
},
{
"epoch": 2.112,
"grad_norm": 28.10418128967285,
"learning_rate": 4.3803418803418805e-05,
"loss": 0.243,
"step": 660
},
{
"epoch": 2.144,
"grad_norm": 9.574631690979004,
"learning_rate": 4.362535612535612e-05,
"loss": 0.2286,
"step": 670
},
{
"epoch": 2.176,
"grad_norm": 14.983197212219238,
"learning_rate": 4.344729344729345e-05,
"loss": 0.2173,
"step": 680
},
{
"epoch": 2.208,
"grad_norm": 19.074600219726562,
"learning_rate": 4.326923076923077e-05,
"loss": 0.2339,
"step": 690
},
{
"epoch": 2.24,
"grad_norm": 27.23953628540039,
"learning_rate": 4.309116809116809e-05,
"loss": 0.2008,
"step": 700
},
{
"epoch": 2.2720000000000002,
"grad_norm": 13.12559986114502,
"learning_rate": 4.291310541310542e-05,
"loss": 0.1996,
"step": 710
},
{
"epoch": 2.304,
"grad_norm": 22.73211097717285,
"learning_rate": 4.2735042735042735e-05,
"loss": 0.1986,
"step": 720
},
{
"epoch": 2.336,
"grad_norm": 12.57758903503418,
"learning_rate": 4.255698005698006e-05,
"loss": 0.2283,
"step": 730
},
{
"epoch": 2.368,
"grad_norm": 61.22172927856445,
"learning_rate": 4.2378917378917385e-05,
"loss": 0.2175,
"step": 740
},
{
"epoch": 2.4,
"grad_norm": 14.797411918640137,
"learning_rate": 4.2200854700854704e-05,
"loss": 0.1758,
"step": 750
},
{
"epoch": 2.432,
"grad_norm": 9.504317283630371,
"learning_rate": 4.202279202279202e-05,
"loss": 0.2087,
"step": 760
},
{
"epoch": 2.464,
"grad_norm": 8.436298370361328,
"learning_rate": 4.184472934472935e-05,
"loss": 0.2318,
"step": 770
},
{
"epoch": 2.496,
"grad_norm": 31.693523406982422,
"learning_rate": 4.166666666666667e-05,
"loss": 0.1909,
"step": 780
},
{
"epoch": 2.528,
"grad_norm": 14.50406265258789,
"learning_rate": 4.148860398860399e-05,
"loss": 0.2157,
"step": 790
},
{
"epoch": 2.56,
"grad_norm": 25.511077880859375,
"learning_rate": 4.131054131054131e-05,
"loss": 0.2237,
"step": 800
},
{
"epoch": 2.592,
"grad_norm": 6.311196804046631,
"learning_rate": 4.1132478632478634e-05,
"loss": 0.1922,
"step": 810
},
{
"epoch": 2.624,
"grad_norm": 37.99801254272461,
"learning_rate": 4.095441595441596e-05,
"loss": 0.1835,
"step": 820
},
{
"epoch": 2.656,
"grad_norm": 48.04703140258789,
"learning_rate": 4.077635327635328e-05,
"loss": 0.191,
"step": 830
},
{
"epoch": 2.6879999999999997,
"grad_norm": 52.908992767333984,
"learning_rate": 4.05982905982906e-05,
"loss": 0.2908,
"step": 840
},
{
"epoch": 2.7199999999999998,
"grad_norm": 8.987102508544922,
"learning_rate": 4.042022792022792e-05,
"loss": 0.216,
"step": 850
},
{
"epoch": 2.752,
"grad_norm": 44.67879104614258,
"learning_rate": 4.024216524216524e-05,
"loss": 0.1991,
"step": 860
},
{
"epoch": 2.784,
"grad_norm": 23.197927474975586,
"learning_rate": 4.006410256410257e-05,
"loss": 0.1751,
"step": 870
},
{
"epoch": 2.816,
"grad_norm": 29.43416404724121,
"learning_rate": 3.988603988603989e-05,
"loss": 0.2064,
"step": 880
},
{
"epoch": 2.848,
"grad_norm": 10.251945495605469,
"learning_rate": 3.970797720797721e-05,
"loss": 0.178,
"step": 890
},
{
"epoch": 2.88,
"grad_norm": 29.398643493652344,
"learning_rate": 3.952991452991453e-05,
"loss": 0.1813,
"step": 900
},
{
"epoch": 2.912,
"grad_norm": 37.31789016723633,
"learning_rate": 3.935185185185186e-05,
"loss": 0.1778,
"step": 910
},
{
"epoch": 2.944,
"grad_norm": 21.298931121826172,
"learning_rate": 3.9173789173789176e-05,
"loss": 0.2124,
"step": 920
},
{
"epoch": 2.976,
"grad_norm": 10.21948528289795,
"learning_rate": 3.89957264957265e-05,
"loss": 0.1584,
"step": 930
},
{
"epoch": 2.9984,
"eval_accuracy": 0.8018779342723005,
"eval_loss": 0.560497522354126,
"eval_runtime": 14.76,
"eval_samples_per_second": 72.155,
"eval_steps_per_second": 2.304,
"step": 937
},
{
"epoch": 3.008,
"grad_norm": 9.502830505371094,
"learning_rate": 3.881766381766382e-05,
"loss": 0.1569,
"step": 940
},
{
"epoch": 3.04,
"grad_norm": 23.640422821044922,
"learning_rate": 3.863960113960114e-05,
"loss": 0.2172,
"step": 950
},
{
"epoch": 3.072,
"grad_norm": 7.1912312507629395,
"learning_rate": 3.846153846153846e-05,
"loss": 0.1954,
"step": 960
},
{
"epoch": 3.104,
"grad_norm": 61.15785598754883,
"learning_rate": 3.828347578347579e-05,
"loss": 0.2364,
"step": 970
},
{
"epoch": 3.136,
"grad_norm": 11.975686073303223,
"learning_rate": 3.8105413105413106e-05,
"loss": 0.1856,
"step": 980
},
{
"epoch": 3.168,
"grad_norm": 13.5460844039917,
"learning_rate": 3.7927350427350425e-05,
"loss": 0.1608,
"step": 990
},
{
"epoch": 3.2,
"grad_norm": 13.04128646850586,
"learning_rate": 3.774928774928775e-05,
"loss": 0.1485,
"step": 1000
},
{
"epoch": 3.232,
"grad_norm": 17.87329864501953,
"learning_rate": 3.7571225071225075e-05,
"loss": 0.1782,
"step": 1010
},
{
"epoch": 3.2640000000000002,
"grad_norm": 21.913515090942383,
"learning_rate": 3.739316239316239e-05,
"loss": 0.1636,
"step": 1020
},
{
"epoch": 3.296,
"grad_norm": 53.9290771484375,
"learning_rate": 3.721509971509972e-05,
"loss": 0.1869,
"step": 1030
},
{
"epoch": 3.328,
"grad_norm": 12.648904800415039,
"learning_rate": 3.7037037037037037e-05,
"loss": 0.1374,
"step": 1040
},
{
"epoch": 3.36,
"grad_norm": 30.469200134277344,
"learning_rate": 3.685897435897436e-05,
"loss": 0.1677,
"step": 1050
},
{
"epoch": 3.392,
"grad_norm": 53.974571228027344,
"learning_rate": 3.668091168091169e-05,
"loss": 0.1297,
"step": 1060
},
{
"epoch": 3.424,
"grad_norm": 13.038790702819824,
"learning_rate": 3.6502849002849005e-05,
"loss": 0.1773,
"step": 1070
},
{
"epoch": 3.456,
"grad_norm": 10.653145790100098,
"learning_rate": 3.6324786324786323e-05,
"loss": 0.1672,
"step": 1080
},
{
"epoch": 3.488,
"grad_norm": 19.27906608581543,
"learning_rate": 3.614672364672365e-05,
"loss": 0.1587,
"step": 1090
},
{
"epoch": 3.52,
"grad_norm": 9.866512298583984,
"learning_rate": 3.5968660968660974e-05,
"loss": 0.1566,
"step": 1100
},
{
"epoch": 3.552,
"grad_norm": 24.01140785217285,
"learning_rate": 3.579059829059829e-05,
"loss": 0.1546,
"step": 1110
},
{
"epoch": 3.584,
"grad_norm": 33.78230285644531,
"learning_rate": 3.561253561253561e-05,
"loss": 0.1769,
"step": 1120
},
{
"epoch": 3.616,
"grad_norm": 29.797826766967773,
"learning_rate": 3.5434472934472935e-05,
"loss": 0.1477,
"step": 1130
},
{
"epoch": 3.648,
"grad_norm": 11.370080947875977,
"learning_rate": 3.525641025641026e-05,
"loss": 0.1749,
"step": 1140
},
{
"epoch": 3.68,
"grad_norm": 6.187923908233643,
"learning_rate": 3.507834757834758e-05,
"loss": 0.1187,
"step": 1150
},
{
"epoch": 3.7119999999999997,
"grad_norm": 57.243919372558594,
"learning_rate": 3.4900284900284904e-05,
"loss": 0.1455,
"step": 1160
},
{
"epoch": 3.7439999999999998,
"grad_norm": 48.05120849609375,
"learning_rate": 3.472222222222222e-05,
"loss": 0.1792,
"step": 1170
},
{
"epoch": 3.776,
"grad_norm": 18.508798599243164,
"learning_rate": 3.454415954415954e-05,
"loss": 0.1977,
"step": 1180
},
{
"epoch": 3.808,
"grad_norm": 11.206692695617676,
"learning_rate": 3.436609686609687e-05,
"loss": 0.1752,
"step": 1190
},
{
"epoch": 3.84,
"grad_norm": 8.796914100646973,
"learning_rate": 3.418803418803419e-05,
"loss": 0.134,
"step": 1200
},
{
"epoch": 3.872,
"grad_norm": 28.23493003845215,
"learning_rate": 3.400997150997151e-05,
"loss": 0.1466,
"step": 1210
},
{
"epoch": 3.904,
"grad_norm": 24.789331436157227,
"learning_rate": 3.3831908831908834e-05,
"loss": 0.1674,
"step": 1220
},
{
"epoch": 3.936,
"grad_norm": 29.97929573059082,
"learning_rate": 3.365384615384616e-05,
"loss": 0.1626,
"step": 1230
},
{
"epoch": 3.968,
"grad_norm": 12.732221603393555,
"learning_rate": 3.347578347578348e-05,
"loss": 0.1162,
"step": 1240
},
{
"epoch": 4.0,
"grad_norm": 23.768869400024414,
"learning_rate": 3.32977207977208e-05,
"loss": 0.1382,
"step": 1250
},
{
"epoch": 4.0,
"eval_accuracy": 0.8497652582159625,
"eval_loss": 0.429149866104126,
"eval_runtime": 14.7304,
"eval_samples_per_second": 72.299,
"eval_steps_per_second": 2.308,
"step": 1250
},
{
"epoch": 4.032,
"grad_norm": 26.3099308013916,
"learning_rate": 3.311965811965812e-05,
"loss": 0.1279,
"step": 1260
},
{
"epoch": 4.064,
"grad_norm": 26.10764503479004,
"learning_rate": 3.294159544159544e-05,
"loss": 0.1868,
"step": 1270
},
{
"epoch": 4.096,
"grad_norm": 42.908058166503906,
"learning_rate": 3.2763532763532764e-05,
"loss": 0.1102,
"step": 1280
},
{
"epoch": 4.128,
"grad_norm": 18.12682342529297,
"learning_rate": 3.258547008547009e-05,
"loss": 0.1372,
"step": 1290
},
{
"epoch": 4.16,
"grad_norm": 32.7211799621582,
"learning_rate": 3.240740740740741e-05,
"loss": 0.1342,
"step": 1300
},
{
"epoch": 4.192,
"grad_norm": 43.664955139160156,
"learning_rate": 3.2229344729344726e-05,
"loss": 0.1205,
"step": 1310
},
{
"epoch": 4.224,
"grad_norm": 21.014827728271484,
"learning_rate": 3.205128205128206e-05,
"loss": 0.1328,
"step": 1320
},
{
"epoch": 4.256,
"grad_norm": 9.477145195007324,
"learning_rate": 3.1873219373219376e-05,
"loss": 0.12,
"step": 1330
},
{
"epoch": 4.288,
"grad_norm": 11.2942533493042,
"learning_rate": 3.1695156695156695e-05,
"loss": 0.1033,
"step": 1340
},
{
"epoch": 4.32,
"grad_norm": 19.64799690246582,
"learning_rate": 3.151709401709402e-05,
"loss": 0.1157,
"step": 1350
},
{
"epoch": 4.352,
"grad_norm": 26.556772232055664,
"learning_rate": 3.133903133903134e-05,
"loss": 0.1053,
"step": 1360
},
{
"epoch": 4.384,
"grad_norm": 17.614032745361328,
"learning_rate": 3.116096866096866e-05,
"loss": 0.1266,
"step": 1370
},
{
"epoch": 4.416,
"grad_norm": 6.941921710968018,
"learning_rate": 3.098290598290599e-05,
"loss": 0.1177,
"step": 1380
},
{
"epoch": 4.448,
"grad_norm": 12.218695640563965,
"learning_rate": 3.080484330484331e-05,
"loss": 0.1209,
"step": 1390
},
{
"epoch": 4.48,
"grad_norm": 43.25284194946289,
"learning_rate": 3.0626780626780625e-05,
"loss": 0.1346,
"step": 1400
},
{
"epoch": 4.5120000000000005,
"grad_norm": 13.644841194152832,
"learning_rate": 3.0448717948717947e-05,
"loss": 0.1156,
"step": 1410
},
{
"epoch": 4.5440000000000005,
"grad_norm": 66.71968078613281,
"learning_rate": 3.0270655270655275e-05,
"loss": 0.1197,
"step": 1420
},
{
"epoch": 4.576,
"grad_norm": 59.87635040283203,
"learning_rate": 3.0092592592592593e-05,
"loss": 0.1105,
"step": 1430
},
{
"epoch": 4.608,
"grad_norm": 10.875226020812988,
"learning_rate": 2.9914529914529915e-05,
"loss": 0.0932,
"step": 1440
},
{
"epoch": 4.64,
"grad_norm": 21.233959197998047,
"learning_rate": 2.9736467236467237e-05,
"loss": 0.0914,
"step": 1450
},
{
"epoch": 4.672,
"grad_norm": 25.02607536315918,
"learning_rate": 2.9558404558404562e-05,
"loss": 0.1175,
"step": 1460
},
{
"epoch": 4.704,
"grad_norm": 9.968689918518066,
"learning_rate": 2.9380341880341884e-05,
"loss": 0.1111,
"step": 1470
},
{
"epoch": 4.736,
"grad_norm": 19.770673751831055,
"learning_rate": 2.9202279202279202e-05,
"loss": 0.1627,
"step": 1480
},
{
"epoch": 4.768,
"grad_norm": 16.976701736450195,
"learning_rate": 2.9024216524216524e-05,
"loss": 0.1355,
"step": 1490
},
{
"epoch": 4.8,
"grad_norm": 32.00942611694336,
"learning_rate": 2.8846153846153845e-05,
"loss": 0.1339,
"step": 1500
},
{
"epoch": 4.832,
"grad_norm": 12.106451034545898,
"learning_rate": 2.866809116809117e-05,
"loss": 0.1192,
"step": 1510
},
{
"epoch": 4.864,
"grad_norm": 20.953189849853516,
"learning_rate": 2.8490028490028492e-05,
"loss": 0.1295,
"step": 1520
},
{
"epoch": 4.896,
"grad_norm": 18.536453247070312,
"learning_rate": 2.8311965811965814e-05,
"loss": 0.101,
"step": 1530
},
{
"epoch": 4.928,
"grad_norm": 14.893009185791016,
"learning_rate": 2.8133903133903132e-05,
"loss": 0.112,
"step": 1540
},
{
"epoch": 4.96,
"grad_norm": 10.388225555419922,
"learning_rate": 2.795584045584046e-05,
"loss": 0.1166,
"step": 1550
},
{
"epoch": 4.992,
"grad_norm": 15.933597564697266,
"learning_rate": 2.777777777777778e-05,
"loss": 0.1058,
"step": 1560
},
{
"epoch": 4.9984,
"eval_accuracy": 0.8666666666666667,
"eval_loss": 0.39110323786735535,
"eval_runtime": 14.8044,
"eval_samples_per_second": 71.938,
"eval_steps_per_second": 2.297,
"step": 1562
},
{
"epoch": 5.024,
"grad_norm": 21.4670352935791,
"learning_rate": 2.75997150997151e-05,
"loss": 0.1265,
"step": 1570
},
{
"epoch": 5.056,
"grad_norm": 22.757436752319336,
"learning_rate": 2.7421652421652423e-05,
"loss": 0.1,
"step": 1580
},
{
"epoch": 5.088,
"grad_norm": 11.63531494140625,
"learning_rate": 2.724358974358974e-05,
"loss": 0.1181,
"step": 1590
},
{
"epoch": 5.12,
"grad_norm": 11.879034996032715,
"learning_rate": 2.706552706552707e-05,
"loss": 0.1001,
"step": 1600
},
{
"epoch": 5.152,
"grad_norm": 13.50586223602295,
"learning_rate": 2.688746438746439e-05,
"loss": 0.098,
"step": 1610
},
{
"epoch": 5.184,
"grad_norm": 6.154415130615234,
"learning_rate": 2.670940170940171e-05,
"loss": 0.1005,
"step": 1620
},
{
"epoch": 5.216,
"grad_norm": 8.229863166809082,
"learning_rate": 2.653133903133903e-05,
"loss": 0.1044,
"step": 1630
},
{
"epoch": 5.248,
"grad_norm": 18.829721450805664,
"learning_rate": 2.6353276353276356e-05,
"loss": 0.1118,
"step": 1640
},
{
"epoch": 5.28,
"grad_norm": 17.462610244750977,
"learning_rate": 2.6175213675213678e-05,
"loss": 0.0809,
"step": 1650
},
{
"epoch": 5.312,
"grad_norm": 15.725268363952637,
"learning_rate": 2.5997150997151e-05,
"loss": 0.109,
"step": 1660
},
{
"epoch": 5.344,
"grad_norm": 12.376166343688965,
"learning_rate": 2.5819088319088318e-05,
"loss": 0.0775,
"step": 1670
},
{
"epoch": 5.376,
"grad_norm": 9.472005844116211,
"learning_rate": 2.564102564102564e-05,
"loss": 0.0958,
"step": 1680
},
{
"epoch": 5.408,
"grad_norm": 16.398996353149414,
"learning_rate": 2.5462962962962965e-05,
"loss": 0.0818,
"step": 1690
},
{
"epoch": 5.44,
"grad_norm": 15.111603736877441,
"learning_rate": 2.5284900284900286e-05,
"loss": 0.0945,
"step": 1700
},
{
"epoch": 5.4719999999999995,
"grad_norm": 13.802181243896484,
"learning_rate": 2.5106837606837608e-05,
"loss": 0.0694,
"step": 1710
},
{
"epoch": 5.504,
"grad_norm": 6.6055216789245605,
"learning_rate": 2.492877492877493e-05,
"loss": 0.0886,
"step": 1720
},
{
"epoch": 5.536,
"grad_norm": 17.351755142211914,
"learning_rate": 2.475071225071225e-05,
"loss": 0.0939,
"step": 1730
},
{
"epoch": 5.568,
"grad_norm": 14.629916191101074,
"learning_rate": 2.4572649572649573e-05,
"loss": 0.0749,
"step": 1740
},
{
"epoch": 5.6,
"grad_norm": 13.987975120544434,
"learning_rate": 2.4394586894586895e-05,
"loss": 0.0696,
"step": 1750
},
{
"epoch": 5.632,
"grad_norm": 19.911739349365234,
"learning_rate": 2.4216524216524217e-05,
"loss": 0.0829,
"step": 1760
},
{
"epoch": 5.664,
"grad_norm": 11.428616523742676,
"learning_rate": 2.4038461538461542e-05,
"loss": 0.0816,
"step": 1770
},
{
"epoch": 5.696,
"grad_norm": 24.872535705566406,
"learning_rate": 2.386039886039886e-05,
"loss": 0.0876,
"step": 1780
},
{
"epoch": 5.728,
"grad_norm": 6.562127113342285,
"learning_rate": 2.3682336182336185e-05,
"loss": 0.0869,
"step": 1790
},
{
"epoch": 5.76,
"grad_norm": 14.26649284362793,
"learning_rate": 2.3504273504273504e-05,
"loss": 0.1098,
"step": 1800
},
{
"epoch": 5.792,
"grad_norm": 10.148120880126953,
"learning_rate": 2.332621082621083e-05,
"loss": 0.0725,
"step": 1810
},
{
"epoch": 5.824,
"grad_norm": 31.81497573852539,
"learning_rate": 2.314814814814815e-05,
"loss": 0.0599,
"step": 1820
},
{
"epoch": 5.856,
"grad_norm": 20.058345794677734,
"learning_rate": 2.297008547008547e-05,
"loss": 0.0987,
"step": 1830
},
{
"epoch": 5.888,
"grad_norm": 23.388731002807617,
"learning_rate": 2.2792022792022794e-05,
"loss": 0.0993,
"step": 1840
},
{
"epoch": 5.92,
"grad_norm": 12.417990684509277,
"learning_rate": 2.2613960113960116e-05,
"loss": 0.0997,
"step": 1850
},
{
"epoch": 5.952,
"grad_norm": 18.029085159301758,
"learning_rate": 2.2435897435897437e-05,
"loss": 0.0875,
"step": 1860
},
{
"epoch": 5.984,
"grad_norm": 8.003488540649414,
"learning_rate": 2.225783475783476e-05,
"loss": 0.0703,
"step": 1870
},
{
"epoch": 6.0,
"eval_accuracy": 0.8028169014084507,
"eval_loss": 0.8593423366546631,
"eval_runtime": 26.4195,
"eval_samples_per_second": 40.311,
"eval_steps_per_second": 1.287,
"step": 1875
},
{
"epoch": 6.016,
"grad_norm": 13.387948036193848,
"learning_rate": 2.207977207977208e-05,
"loss": 0.0723,
"step": 1880
},
{
"epoch": 6.048,
"grad_norm": 3.795748472213745,
"learning_rate": 2.1901709401709402e-05,
"loss": 0.067,
"step": 1890
},
{
"epoch": 6.08,
"grad_norm": 15.893839836120605,
"learning_rate": 2.1723646723646724e-05,
"loss": 0.0718,
"step": 1900
},
{
"epoch": 6.112,
"grad_norm": 6.81917142868042,
"learning_rate": 2.1545584045584046e-05,
"loss": 0.0738,
"step": 1910
},
{
"epoch": 6.144,
"grad_norm": 16.609046936035156,
"learning_rate": 2.1367521367521368e-05,
"loss": 0.0698,
"step": 1920
},
{
"epoch": 6.176,
"grad_norm": 21.796428680419922,
"learning_rate": 2.1189458689458693e-05,
"loss": 0.0681,
"step": 1930
},
{
"epoch": 6.208,
"grad_norm": 9.367144584655762,
"learning_rate": 2.101139601139601e-05,
"loss": 0.0776,
"step": 1940
},
{
"epoch": 6.24,
"grad_norm": 45.68296432495117,
"learning_rate": 2.0833333333333336e-05,
"loss": 0.075,
"step": 1950
},
{
"epoch": 6.272,
"grad_norm": 22.519546508789062,
"learning_rate": 2.0655270655270654e-05,
"loss": 0.1017,
"step": 1960
},
{
"epoch": 6.304,
"grad_norm": 26.589696884155273,
"learning_rate": 2.047720797720798e-05,
"loss": 0.0699,
"step": 1970
},
{
"epoch": 6.336,
"grad_norm": 10.323519706726074,
"learning_rate": 2.02991452991453e-05,
"loss": 0.0628,
"step": 1980
},
{
"epoch": 6.368,
"grad_norm": 19.11711311340332,
"learning_rate": 2.012108262108262e-05,
"loss": 0.0812,
"step": 1990
},
{
"epoch": 6.4,
"grad_norm": 33.67527770996094,
"learning_rate": 1.9943019943019945e-05,
"loss": 0.085,
"step": 2000
},
{
"epoch": 6.432,
"grad_norm": 11.990752220153809,
"learning_rate": 1.9764957264957266e-05,
"loss": 0.0702,
"step": 2010
},
{
"epoch": 6.464,
"grad_norm": 30.36664581298828,
"learning_rate": 1.9586894586894588e-05,
"loss": 0.0628,
"step": 2020
},
{
"epoch": 6.496,
"grad_norm": 21.322391510009766,
"learning_rate": 1.940883190883191e-05,
"loss": 0.0719,
"step": 2030
},
{
"epoch": 6.5280000000000005,
"grad_norm": 29.107898712158203,
"learning_rate": 1.923076923076923e-05,
"loss": 0.0525,
"step": 2040
},
{
"epoch": 6.5600000000000005,
"grad_norm": 17.11159324645996,
"learning_rate": 1.9052706552706553e-05,
"loss": 0.0687,
"step": 2050
},
{
"epoch": 6.592,
"grad_norm": 12.946314811706543,
"learning_rate": 1.8874643874643875e-05,
"loss": 0.0867,
"step": 2060
},
{
"epoch": 6.624,
"grad_norm": 7.321907997131348,
"learning_rate": 1.8696581196581197e-05,
"loss": 0.0661,
"step": 2070
},
{
"epoch": 6.656,
"grad_norm": 34.47880172729492,
"learning_rate": 1.8518518518518518e-05,
"loss": 0.0594,
"step": 2080
},
{
"epoch": 6.688,
"grad_norm": 10.502516746520996,
"learning_rate": 1.8340455840455843e-05,
"loss": 0.0545,
"step": 2090
},
{
"epoch": 6.72,
"grad_norm": 24.073810577392578,
"learning_rate": 1.8162393162393162e-05,
"loss": 0.0592,
"step": 2100
},
{
"epoch": 6.752,
"grad_norm": 25.363536834716797,
"learning_rate": 1.7984330484330487e-05,
"loss": 0.0767,
"step": 2110
},
{
"epoch": 6.784,
"grad_norm": 15.268250465393066,
"learning_rate": 1.7806267806267805e-05,
"loss": 0.069,
"step": 2120
},
{
"epoch": 6.816,
"grad_norm": 8.48227596282959,
"learning_rate": 1.762820512820513e-05,
"loss": 0.0625,
"step": 2130
},
{
"epoch": 6.848,
"grad_norm": 10.944808006286621,
"learning_rate": 1.7450142450142452e-05,
"loss": 0.0729,
"step": 2140
},
{
"epoch": 6.88,
"grad_norm": 12.689846992492676,
"learning_rate": 1.727207977207977e-05,
"loss": 0.0634,
"step": 2150
},
{
"epoch": 6.912,
"grad_norm": 6.600920677185059,
"learning_rate": 1.7094017094017095e-05,
"loss": 0.0758,
"step": 2160
},
{
"epoch": 6.944,
"grad_norm": 19.84024429321289,
"learning_rate": 1.6915954415954417e-05,
"loss": 0.0565,
"step": 2170
},
{
"epoch": 6.976,
"grad_norm": 8.632323265075684,
"learning_rate": 1.673789173789174e-05,
"loss": 0.0671,
"step": 2180
},
{
"epoch": 6.9984,
"eval_accuracy": 0.8892018779342723,
"eval_loss": 0.3630879521369934,
"eval_runtime": 14.8354,
"eval_samples_per_second": 71.788,
"eval_steps_per_second": 2.292,
"step": 2187
},
{
"epoch": 7.008,
"grad_norm": 7.642409324645996,
"learning_rate": 1.655982905982906e-05,
"loss": 0.0489,
"step": 2190
},
{
"epoch": 7.04,
"grad_norm": 8.677549362182617,
"learning_rate": 1.6381766381766382e-05,
"loss": 0.0596,
"step": 2200
},
{
"epoch": 7.072,
"grad_norm": 20.24688148498535,
"learning_rate": 1.6203703703703704e-05,
"loss": 0.0546,
"step": 2210
},
{
"epoch": 7.104,
"grad_norm": 19.509143829345703,
"learning_rate": 1.602564102564103e-05,
"loss": 0.0581,
"step": 2220
},
{
"epoch": 7.136,
"grad_norm": 26.461017608642578,
"learning_rate": 1.5847578347578347e-05,
"loss": 0.0857,
"step": 2230
},
{
"epoch": 7.168,
"grad_norm": 14.442258834838867,
"learning_rate": 1.566951566951567e-05,
"loss": 0.0688,
"step": 2240
},
{
"epoch": 7.2,
"grad_norm": 8.91716194152832,
"learning_rate": 1.5491452991452994e-05,
"loss": 0.079,
"step": 2250
},
{
"epoch": 7.232,
"grad_norm": 15.394043922424316,
"learning_rate": 1.5313390313390312e-05,
"loss": 0.0561,
"step": 2260
},
{
"epoch": 7.264,
"grad_norm": 19.662343978881836,
"learning_rate": 1.5135327635327638e-05,
"loss": 0.0581,
"step": 2270
},
{
"epoch": 7.296,
"grad_norm": 8.76469898223877,
"learning_rate": 1.4957264957264958e-05,
"loss": 0.0439,
"step": 2280
},
{
"epoch": 7.328,
"grad_norm": 10.02730941772461,
"learning_rate": 1.4779202279202281e-05,
"loss": 0.0547,
"step": 2290
},
{
"epoch": 7.36,
"grad_norm": 21.142120361328125,
"learning_rate": 1.4601139601139601e-05,
"loss": 0.0618,
"step": 2300
},
{
"epoch": 7.392,
"grad_norm": 8.849529266357422,
"learning_rate": 1.4423076923076923e-05,
"loss": 0.0601,
"step": 2310
},
{
"epoch": 7.424,
"grad_norm": 13.313969612121582,
"learning_rate": 1.4245014245014246e-05,
"loss": 0.0578,
"step": 2320
},
{
"epoch": 7.456,
"grad_norm": 17.729047775268555,
"learning_rate": 1.4066951566951566e-05,
"loss": 0.0593,
"step": 2330
},
{
"epoch": 7.4879999999999995,
"grad_norm": 13.839546203613281,
"learning_rate": 1.388888888888889e-05,
"loss": 0.0566,
"step": 2340
},
{
"epoch": 7.52,
"grad_norm": 6.299365997314453,
"learning_rate": 1.3710826210826211e-05,
"loss": 0.0469,
"step": 2350
},
{
"epoch": 7.552,
"grad_norm": 12.264620780944824,
"learning_rate": 1.3532763532763535e-05,
"loss": 0.0476,
"step": 2360
},
{
"epoch": 7.584,
"grad_norm": 13.175020217895508,
"learning_rate": 1.3354700854700855e-05,
"loss": 0.0483,
"step": 2370
},
{
"epoch": 7.616,
"grad_norm": 8.695143699645996,
"learning_rate": 1.3176638176638178e-05,
"loss": 0.0676,
"step": 2380
},
{
"epoch": 7.648,
"grad_norm": 7.434255599975586,
"learning_rate": 1.29985754985755e-05,
"loss": 0.0514,
"step": 2390
},
{
"epoch": 7.68,
"grad_norm": 11.564408302307129,
"learning_rate": 1.282051282051282e-05,
"loss": 0.0496,
"step": 2400
},
{
"epoch": 7.712,
"grad_norm": 15.317635536193848,
"learning_rate": 1.2642450142450143e-05,
"loss": 0.0723,
"step": 2410
},
{
"epoch": 7.744,
"grad_norm": 5.177554130554199,
"learning_rate": 1.2464387464387465e-05,
"loss": 0.0508,
"step": 2420
},
{
"epoch": 7.776,
"grad_norm": 21.86553382873535,
"learning_rate": 1.2286324786324787e-05,
"loss": 0.0531,
"step": 2430
},
{
"epoch": 7.808,
"grad_norm": 5.066462516784668,
"learning_rate": 1.2108262108262108e-05,
"loss": 0.0601,
"step": 2440
},
{
"epoch": 7.84,
"grad_norm": 3.860797643661499,
"learning_rate": 1.193019943019943e-05,
"loss": 0.0529,
"step": 2450
},
{
"epoch": 7.872,
"grad_norm": 27.842437744140625,
"learning_rate": 1.1752136752136752e-05,
"loss": 0.0736,
"step": 2460
},
{
"epoch": 7.904,
"grad_norm": 26.736413955688477,
"learning_rate": 1.1574074074074075e-05,
"loss": 0.0589,
"step": 2470
},
{
"epoch": 7.936,
"grad_norm": 25.008304595947266,
"learning_rate": 1.1396011396011397e-05,
"loss": 0.0549,
"step": 2480
},
{
"epoch": 7.968,
"grad_norm": 4.6437506675720215,
"learning_rate": 1.1217948717948719e-05,
"loss": 0.0533,
"step": 2490
},
{
"epoch": 8.0,
"grad_norm": 26.05324935913086,
"learning_rate": 1.103988603988604e-05,
"loss": 0.0457,
"step": 2500
},
{
"epoch": 8.0,
"eval_accuracy": 0.8338028169014085,
"eval_loss": 0.7270519137382507,
"eval_runtime": 16.0951,
"eval_samples_per_second": 66.169,
"eval_steps_per_second": 2.112,
"step": 2500
},
{
"epoch": 8.032,
"grad_norm": 6.876763820648193,
"learning_rate": 1.0861823361823362e-05,
"loss": 0.0392,
"step": 2510
},
{
"epoch": 8.064,
"grad_norm": 9.069485664367676,
"learning_rate": 1.0683760683760684e-05,
"loss": 0.059,
"step": 2520
},
{
"epoch": 8.096,
"grad_norm": 14.368523597717285,
"learning_rate": 1.0505698005698005e-05,
"loss": 0.0455,
"step": 2530
},
{
"epoch": 8.128,
"grad_norm": 10.847954750061035,
"learning_rate": 1.0327635327635327e-05,
"loss": 0.0368,
"step": 2540
},
{
"epoch": 8.16,
"grad_norm": 7.805965900421143,
"learning_rate": 1.014957264957265e-05,
"loss": 0.0374,
"step": 2550
},
{
"epoch": 8.192,
"grad_norm": 37.152957916259766,
"learning_rate": 9.971509971509972e-06,
"loss": 0.0434,
"step": 2560
},
{
"epoch": 8.224,
"grad_norm": 32.26515197753906,
"learning_rate": 9.793447293447294e-06,
"loss": 0.0538,
"step": 2570
},
{
"epoch": 8.256,
"grad_norm": 9.871603012084961,
"learning_rate": 9.615384615384616e-06,
"loss": 0.0443,
"step": 2580
},
{
"epoch": 8.288,
"grad_norm": 18.939363479614258,
"learning_rate": 9.437321937321937e-06,
"loss": 0.0444,
"step": 2590
},
{
"epoch": 8.32,
"grad_norm": 8.318560600280762,
"learning_rate": 9.259259259259259e-06,
"loss": 0.0367,
"step": 2600
},
{
"epoch": 8.352,
"grad_norm": 10.871614456176758,
"learning_rate": 9.081196581196581e-06,
"loss": 0.0493,
"step": 2610
},
{
"epoch": 8.384,
"grad_norm": 5.46032190322876,
"learning_rate": 8.903133903133903e-06,
"loss": 0.0413,
"step": 2620
},
{
"epoch": 8.416,
"grad_norm": 1.5709623098373413,
"learning_rate": 8.725071225071226e-06,
"loss": 0.0658,
"step": 2630
},
{
"epoch": 8.448,
"grad_norm": 26.417600631713867,
"learning_rate": 8.547008547008548e-06,
"loss": 0.0681,
"step": 2640
},
{
"epoch": 8.48,
"grad_norm": 9.628387451171875,
"learning_rate": 8.36894586894587e-06,
"loss": 0.0399,
"step": 2650
},
{
"epoch": 8.512,
"grad_norm": 2.8144702911376953,
"learning_rate": 8.190883190883191e-06,
"loss": 0.0435,
"step": 2660
},
{
"epoch": 8.544,
"grad_norm": 13.873322486877441,
"learning_rate": 8.012820512820515e-06,
"loss": 0.0501,
"step": 2670
},
{
"epoch": 8.576,
"grad_norm": 4.96248722076416,
"learning_rate": 7.834757834757835e-06,
"loss": 0.0423,
"step": 2680
},
{
"epoch": 8.608,
"grad_norm": 12.236059188842773,
"learning_rate": 7.656695156695156e-06,
"loss": 0.0488,
"step": 2690
},
{
"epoch": 8.64,
"grad_norm": 6.332564353942871,
"learning_rate": 7.478632478632479e-06,
"loss": 0.0437,
"step": 2700
},
{
"epoch": 8.672,
"grad_norm": 7.448380470275879,
"learning_rate": 7.3005698005698005e-06,
"loss": 0.035,
"step": 2710
},
{
"epoch": 8.704,
"grad_norm": 7.228295803070068,
"learning_rate": 7.122507122507123e-06,
"loss": 0.0457,
"step": 2720
},
{
"epoch": 8.736,
"grad_norm": 5.306349277496338,
"learning_rate": 6.944444444444445e-06,
"loss": 0.0496,
"step": 2730
},
{
"epoch": 8.768,
"grad_norm": 15.959365844726562,
"learning_rate": 6.766381766381767e-06,
"loss": 0.04,
"step": 2740
},
{
"epoch": 8.8,
"grad_norm": 13.594867706298828,
"learning_rate": 6.588319088319089e-06,
"loss": 0.0466,
"step": 2750
},
{
"epoch": 8.832,
"grad_norm": 7.143582820892334,
"learning_rate": 6.41025641025641e-06,
"loss": 0.0367,
"step": 2760
},
{
"epoch": 8.864,
"grad_norm": 10.158649444580078,
"learning_rate": 6.2321937321937325e-06,
"loss": 0.0442,
"step": 2770
},
{
"epoch": 8.896,
"grad_norm": 4.369603633880615,
"learning_rate": 6.054131054131054e-06,
"loss": 0.0352,
"step": 2780
},
{
"epoch": 8.928,
"grad_norm": 11.626177787780762,
"learning_rate": 5.876068376068376e-06,
"loss": 0.0414,
"step": 2790
},
{
"epoch": 8.96,
"grad_norm": 8.576985359191895,
"learning_rate": 5.6980056980056985e-06,
"loss": 0.0445,
"step": 2800
},
{
"epoch": 8.992,
"grad_norm": 11.782782554626465,
"learning_rate": 5.51994301994302e-06,
"loss": 0.0396,
"step": 2810
},
{
"epoch": 8.9984,
"eval_accuracy": 0.8826291079812206,
"eval_loss": 0.46553394198417664,
"eval_runtime": 16.3312,
"eval_samples_per_second": 65.213,
"eval_steps_per_second": 2.082,
"step": 2812
},
{
"epoch": 9.024,
"grad_norm": 9.517723083496094,
"learning_rate": 5.341880341880342e-06,
"loss": 0.0281,
"step": 2820
},
{
"epoch": 9.056,
"grad_norm": 11.273005485534668,
"learning_rate": 5.163817663817664e-06,
"loss": 0.0395,
"step": 2830
},
{
"epoch": 9.088,
"grad_norm": 12.864770889282227,
"learning_rate": 4.985754985754986e-06,
"loss": 0.0399,
"step": 2840
},
{
"epoch": 9.12,
"grad_norm": 11.933670997619629,
"learning_rate": 4.807692307692308e-06,
"loss": 0.0381,
"step": 2850
},
{
"epoch": 9.152,
"grad_norm": 21.274028778076172,
"learning_rate": 4.6296296296296296e-06,
"loss": 0.03,
"step": 2860
},
{
"epoch": 9.184,
"grad_norm": 6.597903251647949,
"learning_rate": 4.451566951566951e-06,
"loss": 0.042,
"step": 2870
},
{
"epoch": 9.216,
"grad_norm": 1.0557407140731812,
"learning_rate": 4.273504273504274e-06,
"loss": 0.0268,
"step": 2880
},
{
"epoch": 9.248,
"grad_norm": 11.865984916687012,
"learning_rate": 4.0954415954415956e-06,
"loss": 0.0301,
"step": 2890
},
{
"epoch": 9.28,
"grad_norm": 25.20126724243164,
"learning_rate": 3.917378917378917e-06,
"loss": 0.0383,
"step": 2900
},
{
"epoch": 9.312,
"grad_norm": 11.134941101074219,
"learning_rate": 3.7393162393162394e-06,
"loss": 0.0375,
"step": 2910
},
{
"epoch": 9.344,
"grad_norm": 2.754523992538452,
"learning_rate": 3.5612535612535615e-06,
"loss": 0.043,
"step": 2920
},
{
"epoch": 9.376,
"grad_norm": 12.286087989807129,
"learning_rate": 3.3831908831908837e-06,
"loss": 0.035,
"step": 2930
},
{
"epoch": 9.408,
"grad_norm": 3.8736507892608643,
"learning_rate": 3.205128205128205e-06,
"loss": 0.0391,
"step": 2940
},
{
"epoch": 9.44,
"grad_norm": 22.06711769104004,
"learning_rate": 3.027065527065527e-06,
"loss": 0.0377,
"step": 2950
},
{
"epoch": 9.472,
"grad_norm": 8.435704231262207,
"learning_rate": 2.8490028490028492e-06,
"loss": 0.0301,
"step": 2960
},
{
"epoch": 9.504,
"grad_norm": 27.129562377929688,
"learning_rate": 2.670940170940171e-06,
"loss": 0.039,
"step": 2970
},
{
"epoch": 9.536,
"grad_norm": 4.00566291809082,
"learning_rate": 2.492877492877493e-06,
"loss": 0.0261,
"step": 2980
},
{
"epoch": 9.568,
"grad_norm": 18.572261810302734,
"learning_rate": 2.3148148148148148e-06,
"loss": 0.036,
"step": 2990
},
{
"epoch": 9.6,
"grad_norm": 6.907130241394043,
"learning_rate": 2.136752136752137e-06,
"loss": 0.0455,
"step": 3000
},
{
"epoch": 9.632,
"grad_norm": 3.8767590522766113,
"learning_rate": 1.9586894586894586e-06,
"loss": 0.0456,
"step": 3010
},
{
"epoch": 9.664,
"grad_norm": 12.119085311889648,
"learning_rate": 1.7806267806267808e-06,
"loss": 0.037,
"step": 3020
},
{
"epoch": 9.696,
"grad_norm": 7.650444507598877,
"learning_rate": 1.6025641025641025e-06,
"loss": 0.0363,
"step": 3030
},
{
"epoch": 9.728,
"grad_norm": 8.467825889587402,
"learning_rate": 1.4245014245014246e-06,
"loss": 0.0198,
"step": 3040
},
{
"epoch": 9.76,
"grad_norm": 19.763225555419922,
"learning_rate": 1.2464387464387465e-06,
"loss": 0.0408,
"step": 3050
},
{
"epoch": 9.792,
"grad_norm": 25.160770416259766,
"learning_rate": 1.0683760683760685e-06,
"loss": 0.0392,
"step": 3060
},
{
"epoch": 9.824,
"grad_norm": 7.482017517089844,
"learning_rate": 8.903133903133904e-07,
"loss": 0.0283,
"step": 3070
},
{
"epoch": 9.856,
"grad_norm": 7.559390544891357,
"learning_rate": 7.122507122507123e-07,
"loss": 0.0463,
"step": 3080
},
{
"epoch": 9.888,
"grad_norm": 10.798991203308105,
"learning_rate": 5.341880341880342e-07,
"loss": 0.0298,
"step": 3090
},
{
"epoch": 9.92,
"grad_norm": 3.9386239051818848,
"learning_rate": 3.5612535612535615e-07,
"loss": 0.0386,
"step": 3100
},
{
"epoch": 9.952,
"grad_norm": 15.766891479492188,
"learning_rate": 1.7806267806267808e-07,
"loss": 0.0277,
"step": 3110
},
{
"epoch": 9.984,
"grad_norm": 3.3344645500183105,
"learning_rate": 0.0,
"loss": 0.0293,
"step": 3120
},
{
"epoch": 9.984,
"eval_accuracy": 0.8610328638497653,
"eval_loss": 0.7580540180206299,
"eval_runtime": 16.5093,
"eval_samples_per_second": 64.509,
"eval_steps_per_second": 2.059,
"step": 3120
},
{
"epoch": 9.984,
"step": 3120,
"total_flos": 9.926487761391452e+18,
"train_loss": 0.15716634974934351,
"train_runtime": 142469.0271,
"train_samples_per_second": 2.808,
"train_steps_per_second": 0.022
}
],
"logging_steps": 10,
"max_steps": 3120,
"num_input_tokens_seen": 0,
"num_train_epochs": 10,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 9.926487761391452e+18,
"train_batch_size": 32,
"trial_name": null,
"trial_params": null
}